Effect of Molecular Architecture of PDMAEMA ... - ACS Publications

Article pubs.acs.org/JPCB

Effect of Molecular Architecture of PDMAEMA−POEGMA Random and Block Copolymers on Their Adsorption on Regenerated and Anionic Nanocelluloses and Evidence of Interfacial Water Expulsion Maija Vuoriluoto,† Hannes Orelma,*,†,‡ Leena-Sisko Johansson,† Baolei Zhu,§ Mikko Poutanen,∥ Andreas Walther,§ Janne Laine,† and Orlando J. Rojas† †

Biobased Colloids and Materials group (BiCMat), Department of Forest Products Technology, School of Chemical Technology, Aalto University, FI-00076, Espoo, Finland ‡ VTT, Technical Research Centre of Finland, Biologinkuja 7, P.O. Box 1000, FIN-02044 VTT, Finland § DWI − Leibniz-Institute for Interactive Materials Research, Forckenbeckstr. 50, D-52056 Aachen, Germany ∥ Department of Applied Physics, School of Science, Aalto University, FI-00076, Espoo, Finland S Supporting Information *

ABSTRACT: Block copolymers of poly(2-(dimethylamino)ethyl methacrylate) (PDMAEMA) and poly(oligo(ethylene glycol) methyl ether methacrylate) (POEGMA) with varying block sizes were synthesized by consecutive reversible addition− fragmentation chain transfer (RAFT) polymerization and then exposed to cellulose substrates with different anionic charge density. The extent and dynamics of quaternized PDMAEMA-b-POEGMA adsorption on regenerated cellulose, cellulose nanofibrils (CNF), and (2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO)-oxidized cellulose nanofibrils (TOCNF) was determined by using electromechanical and optical techniques, namely, quartz crystal microbalance (QCM-D) and surface plasmon resonance (SPR), respectively. PDMAEMA-b-POEGMA equilibrium adsorption increased with the anionic charge of cellulose, an indication of electrostatic interactions. Such an observation was further confirmed by atomic force microscopy (AFM) and X-ray photoelectron spectroscopy (XPS). Depending on their architecture, adsorption on TOCNF of some of the PDMAEMA-b-POEGMA copolymers produced a significant reduction in QCM frequency, as expected from large mass uptake, while surprisingly, other copolymers induced the opposite effect. This latter, remarkable behavior was ascribed to coupled water expulsion from the interface upon charge neutralization of anionic surface sites with adsorbing cationic polymer segments. These observations were further investigated with SPR and QCM-D measurements using deuterium oxide solvent exchange to determine the amount of coupled water at the TOCNF−block copolymer interface. Finally, random copolymers with similar composition adsorbed to a larger extent compared to the respective block copolymers, revealing the effect of adsorbed loops and tails as well as hydration.

■

filters,6 diagnostic systems,7,8 and barrier films.9 CNF is typically extracted from plant biomaterial through mechanical4,10 or chemical11,12 treatments. CNF also displays a large water uptake ability and a tendency to form hydrogels due to physical and chemical cross-linking.11 Therefore, the wide application field of

INTRODUCTION Developing technologies based on renewable materials to support ever increased consumption of carbon-based materials has become increasingly important. Cellulose is a renewable material that can offer solutions to many of these demands. Nanoscale cellulosic materials and specifically cellulose nanofibrils (CNF) have been proposed to improve functionality in many applications as reinforcement agents in composites,1,2 for fiber technologies,3 rheology modifiers,4 absorbent materials,5 © 2015 American Chemical Society

Received: August 6, 2015 Revised: November 3, 2015 Published: November 11, 2015 15275

DOI: 10.1021/acs.jpcb.5b07628 J. Phys. Chem. B 2015, 119, 15275−15286

Article

The Journal of Physical Chemistry B

Scheme 1. Molecular Structure of Quaternized PDMAEMA−POEGMA Copolymers (a) and Schematic Illustration of the Block Copolymer (b) and Random Copolymer (c) Adsorption on Highly Charged TEMPO-Oxidized CNF (TOCNF) Substrate

random copolymers of DMAEMA and OEGMA were investigated here for this purpose. CNF−water interactions have a major influence in many related applications due to high water uptake ability and hydrogel formation. One of the consequences of this behavior is that, unless a judicious change in CNF charge density is applied, only limited dispersion in aqueous media occurs once it has been dried.31 Thus, CNF applications can be highly susceptible depending on its exposure to water, humid air, etc. Relevant to the present investigation are the available ultrathin films of cellulose, which can be studied with surface-sensitive techniques to unveil fundamental aspects of cellulose−water interactions as well as surface (physical and chemical) modification. Interactions of the cellulose polymer with water have already been extensively investigated;32,33 recently the interactions between CNF and water have been at the center of interest given its significance in application development. The swelling of CNF ultrathin films in water34,35 and water uptake of CNF multilayers36 has been reported. Interestingly, it was found that cationic polyelectrolyte adsorption can reduce swelling of CNF in water.35 Therefore, the impact on water interactions upon adsorption on cellulose nanofibrils (CNF) and (2,2,6,6tetramethylpiperidin-1-yl)oxyl (TEMPO)-oxidized CNF (TOCNF) of highly hydrophilic block copolymers containing cationic segments is elucidated in this study. A H2O/D2O solvent exchange method in QCM-D37 was applied for this purpose, as it allows one to determine swelling of regenerated cellulose, cellulose nanocrystals,38 and CNF35 films. To this end, the adsorption of block copolymers containing quaternized PDMAEMA and POEGMA on low-charged regenerated cellulose, medium-charged CNF, and highly charged TOCNF was investigated in situ by quartz crystal microbalance with dissipation (QCM-D) monitoring39 and surface plasmon resonance (SPR).40 QCM-D experiments with H2O/D2O solvent exchange were applied to determine the water content of TOCNF films before and after block copolymer adsorption. The changes in topography and surface chemical composition were determined by atomic force microscopy (AFM) and X-ray photoelectron spectroscopy (XPS), respectively. To study the effect of molecular architecture on adsorption, random copolymers of PDMAEMA and POEGMA were adsorbed on TEMPO-oxidized CNF. Scheme 1 illustrates a schematic

nanocellulose requires a better understanding of its inherent properties and interaction with water, especially if its chemical features are changed via physical or chemical modification. One simple and effective way for altering the surface chemistry of cellulose is through adsorption of functional surfactants and polymers. We have investigated the adsorption on cellulose of nonionic block copolymers13−16 as well as triblock copolymers end-capped with cationic chains.17 Likewise, the attachment on hydrophilic surfaces of copolymers containing cationic poly(2(dimethylamino)ethyl methacrylate) (PDMAEMA) with hydrophobic segments has been reported, and the adsorption mechanism has been attributed to electrostatic interactions between the cationic segments and the anionic surface, both for quaternized18 or nonquaternized19 PDMAEMA. In this work, we study the adsorption on cellulose of similar but entirely hydrophilic copolymers, carrying blocks of positively charged PDMAEMA20 and highly hydrophilic poly(oligo(ethylene glycol) methyl ether methacrylate) (POEGMA). The cationic PDMAEMA block was hypothesized to function as an anchor for POEGMA onto CNF. Noteworthy, polyethylene glycol molecules (PEG) do not have a high affinity with cellulose21 but the attachment of PEG on cellulosic material through various indirect routes has been demonstrated.22,23 CNF modified with block copolymers with PEG-like molecules, which are highly hydrophilic and are in extended conformation in the surrounding aqueous media, find use to stabilize dispersions24 and provide surface passivation toward nonspecific biomolecule adsorption.25 Additionally, PEG molecules can supply lubrication in an aqueous environment26,27 especially relevant to food, pharmaceutical,28 and biomedical29 materials. Recently, multifunctional polymers containing PEG molecules have been demonstrated to function as promising antifouling agents toward proteins.30 Therefore, these entirely hydrophilic block copolymers could potentially be effective in producing antibiofouling cellulose surfaces. Triblock PDMAEMA-b-POEGMA-b-PDMAEMA copolymers may offer interesting cross-linking options compared to block PDMAEMA-b-POEGMA polymers. On the other hand, nonionic symmetric triblock copolymers (such as Pluronics), which are expected to be effective in antifouling, adsorb weakly and only to a limited extent on cellulose.16 To study the effect of molecular architecture on adsorption and layer properties, 15276


Article


diameter of the gel particles measured 10 μm, and the nominal pore widths were 30, 100, 1000, and 3000 Å. Calibration was achieved using narrow distributed poly(methyl methacrylate) standards (Polymer Standards Service). The as-prepared PDMAEMA-b-POEGMA copolymers contained varying DMAEMA and OEGMA repeating units: 33, 58, or 74 for DMAEMA and 10 to 137 for OEGMA. A total of nine block copolymers combining these block sizes were used and referred to as Dn-EGMAm, where n and m represent the number of DMAEMA and OEGMA units in the copolymer, respectively. In addition, four random copolymers of PDMAEMA and POEGMA (PDMAEMA-rnd-POEGMA) with well-defined molar percentage ratios (10−90, 30−70, 70−30, 90−10) were synthesized by ATRP and quaternized by methyl iodide. Preparation of Cellulose Substrates. Regenerated cellulose, cellulose nanofibrils (CNF), and TEMPO-oxidized CNF (TOCNF) were utilized in the form of ultrathin films that were used as adsorbing substrate. All of the cellulose films were stored under ambient conditions and were allowed to stabilize in water overnight prior to measurements in QCM-D or SPR (see methods further below). Regenerated cellulose thin films were deposited on QCM-D silica crystals (Q-Sense, Västra Frölunda, Sweden) from TMSC according to Kontturi et al.42 The QCM-D crystals were first cleaned with UV/ozone treatment for 20 min and subsequently spin-coated (model WS-650SX-6NPP, Laurell Technologies, PA, USA) at 4000 rpm for 60 s with a 10 g/L solution of TMSC in toluene. The TMSC surfaces were hydrolyzed to regenerated cellulose by a hydrochloric acid vapor treatment. CNF thin films were prepared according to Ahola et al.43 In short, CNF gel was diluted (0.190 wt % CNF in water) and tipultrasonicated for 10 min at 25% amplitude setting and consecutively centrifuged at 10400 rpm for 45 min. Individual nanofibrils were obtained from the resultant clear supernatant. The nanofibrils were spin-coated onto UV/ozonized silica QCM-D or gold SPR (Oy BioNavis Ltd., Ylöjärvi, Finland) wafers with a preadsorbed thin anchor layer of PEI at 3000 rpm with 90 s spinning time. After spin-coating, the CNF films were dried at 80 °C for 10 min. CNF films were TEMPO-oxidized in situ in the QCM-D and SPR modules, thereby introducing carboxylic groups on the surface. The oxidation process was monitored in real time in the QCM and SPR setups (see Supporting Information S1 for the typical QCM-D and SPR sensograms upon TEMPO oxidation of CNF films). In short, CNF films were subjected to continuous flow of TEMPO solution (0.13 mmol of TEMPO, 4.65 mmol of NaBr, and 5.00 mmol of NaClO in water, pH 10) for 2 min after which oxidation was stopped by ethanol addition for 2 min. MilliQ-water rinsing was applied to remove any excess oxidation chemicals. Subsequently, the films were allowed to stabilize in 2.5 mM phosphate buffer solution at pH 6.8 (ionic strength 4 mM) and they were used for block copolymer adsorption measurements immediately without drying steps. Quartz Crystal Microbalance with Dissipation (QCMD). The adsorption of the block and random copolymers on regenerated cellulose, CNF, and TOCNF films were monitored with a Q-Sense E4 apparatus (Västra Frölunda, Sweden). The QCM-D technique is based on the acoustic oscillation of a quartz crystal by which adsorption events can be monitored.39 The dissipation values measured simultaneously yield information about the viscoelastic properties of adsorbing layers.44 The frequency and dissipation changes were measured at a fundamental resonance frequency and its overtones. In an ideal

representation of PDMAEMA−POEGMA block and random copolymer structure and adsorption on highly charged TOCNF.

■

EXPERIMENTAL SECTION Materials. Cellulose nanofibrils (CNF) were produced by processing bleached sulphite birch fibers through a Masuko grinder (three passes) after which they were further microfluidized (at least 12 passes) through a M110P fluidizer (Microfluidics Corp., Newton, MA, USA) equipped with 200 and 100 μm chambers and operated at 2000 bar pressure. Trimethylsilylcellulose (TMSC) was synthesized as previously described.41 All used laboratory chemicals were of analytical grade. Deionized water further purified with a Millipore Synergy UV unit (Milli-Q) was used in all experiments. Reversible Addition−Fragmentation Chain Transfer (RAFT) Polymerization of PDMAEMA. A 1.65 g portion of DMAEMA (2-(dimethylamino)ethyl methacrylate, 10.5 mmol) and 14.1 mg (0.51 mmol) of 4-cyano-4-(phenylcarbonothioylthio) pentanoic acid was weighed and dissolved in 8 mL of 1,4dioxane in a container with a stirrer. Afterward, 2.5 mg (0.015 mmol) of 2,2′-azobis(2-methylpropionitrile) was added, and the container was sealed with a rubber septum and bubbled with N2 for 20 min to remove oxygen. The container was placed into an oil bath at 70 °C. Samples were taken during the reaction to monitor conversion and molecular weight evolution. After 15 h, the reaction system was cooled down in an ice bath. The polymer was precipitated in hexane at least three times to remove the unreacted monomer and freeze-dried to yield a pure polymer. PDMAEMA precursors with different molecular weights were made. Chain Extension of PDMAEMA to PDMAEMA-bPOEGMA. The chain extension of PDMAEMA to PDMAEMA-b-POEGMA was carried out in a similar manner as RAFT polymerization of PDMAEMA. A 2.5 g portion of OEGMA (oligo(ethylene glycol) methyl ether methacrylate, 5.3 mmol) and 0.169 g of PDMAEMA precursor (0.0212 mmol) was weighed and dissolved in a container with a stirrer. Afterward, 0.7 mg of AIBN (0.0042 mmol) was added, and the container was sealed with a rubber septum and bubbled with N2 for 20 min to remove oxygen. The container was placed into an oil bath at 65 °C. Samples were taken during the reaction to monitor the conversion and molecular weight. After 7 h, the reaction system was cooled down in an ice bath. The polymers were dialyzed against water for 7 days and freeze-dried to yield the purified polymers. The samples were quaternized by dissolving 1 g of block copolymer in 15 mL of 1,4-dioxane in a vial with a stirrer. Afterward, 0.52 mL of methyl iodide was added into the polymer solution in a quick manner while keeping the solution stirring rigorously. After 24 h, the quaternization reaction was complete. The polymer solution was freeze-dried to get rid of the solvent and residual methyl iodide. A yellowish powder was yielded. Characterization of PDMAEMA-b-POEGMA. 1H NMR spectra (400 MHz) of the polymers were recorded using a Bruker DPX 400 instrument. All spectra were referenced internally to residual proton signals of the deuterated solvent (CDCl3). SEC with DMF (HPLC grade, VWR) as eluent was performed using an Agilent 1100 system equipped with a dual RI-/Visco detector (ETA-2020, WGE). The eluent contained 1 g·L−1 LiBr (≥99%, Sigma-Aldrich). The sample solvent contained traces of distilled water as internal standard. One precolumn (8 × 50 mm) and four GRAM gel columns (8 × 300 mm, Polymer Standards Service) were applied at a flow rate of 1.0 mL·min−1 at 40 °C. The 15277


Article


the angles and wavelengths at which the surface plasmon resonance effect occurs.51,52 Equation 3 was used to determine the thickness of the adsorbed layer51

case, the adsorbed mass is proportional to the change in resonance frequency according to the Sauerbrey equation45,46 Δm = −CQCM ‐ D

Δf n

(1)

d=

where CQCM‑D is 17.7 ng Hz−1 cm−2 for 5 MHz crystal (provided by the manufacturer), Δf is the change in frequency, and n corresponds to the overtone number. The Sauerbrey equation is applicable only if the adsorbed layer is uniformly distributed on the crystal surface, it is rigid, and the mass is small compared to the mass of the crystal; if these conditions are not fulfilled, the calculation results in mass overestimation. Under such conditions, the Voigt viscoelastic model is more appropriate. Therefore, we used the Voigt protocol (Q-Tools software, version 2.1, Q-Sense, Västra Frölunda, Sweden) to estimate the adsorbed mass of block copolymers on the surfaces by assuming a density of 1200 g/m3 for the adsorbed layer.47 All QCM-D experiments were repeated at least twice under a constant flow rate of 100 μL/min at 23 °C. Determination of Coupled Water with H2O/D2O Solvent Exchange in QCM-D. QCM-D was also utilized to determine the mass of spin-coated CNF on the silica sensor crystal by comparing the recorded fundamental resonance frequency of the sensor in air before and after CNF deposition. The water content of CNF and TOCNF films was determined according to a previously published method,37,38 by changing the water solvent to deuterium oxide (D2O) and comparing the scaled QCM-D frequency shift. The contribution of bound water within the film (Δf/n)H2O can be calculated from eq 238,48 ⎛ Δf ⎞ ⎜ ⎟ = ⎝ n ⎠H O 2

Δf n film

( )

−

(3)

where Δangle is the change in the SPR angle, ld is a characteristic evanescent electromagnetic field decay length, estimated as 0.37 of the light wavelength (240 nm), m is a sensitivity factor for the sensor obtained after calibration of the SPR (109.94°/RIU), n0 is the refractive index of the bulk solution (1.334 RIU), and na is the refractive index of the adsorbed substance. The refractive index was estimated as 1.46 for the block copolymers (based on POEGMA).53 The adsorbed amounts per unit area were calculated following eq 452 Δm = d × ρ

(4)

where d is the thickness of the adsorbed layer and ρ is the packing density of the adsorbed species. The assumed packing density was 1.15 g/cm3 based on POEGMA54 for the block copolymers. All SPR measurements were conducted with a 100 μL/min flow rate at 20 °C. Each SPR measurement was performed at least twice. Adsorption of Block Copolymers on Cellulose Films. The adsorption of nine block copolymers with varying block sizes on regenerated cellulose, CNF, and TOCNF was studied with QCM-D and SPR. The block copolymers (0.5 g/L concentration) were adsorbed on the cellulose surfaces from 2.5 mM phosphate buffer solution at pH 6.8 until an adsorption plateau was observed. Upon reaching equilibrium, rinsing was applied with the respective buffer solution. Adsorption of Random Copolymers on TOCNF Films. Nonquaternized and quaternized random copolymers with given PDMAEMA-to-POEGMA ratios (10−90%, 30−70%, 70−30%, 90−10%) were adsorbed onto TOCNF films, and the adsorption was monitored via SPR. The nonquaternized versions of the random copolymers were adsorbed onto TOCNF from pH 5.5 acetate buffer solution (10 mM) and from pH 9.5 carbonate buffer solution (10 mM). Likewise, the quaternized polymers were applied onto TOCNF from neutral pH, Milli-Q-water. Rinsing was applied with the respective (background) solvent once the adsorption plateau was reached for the nonquaternized samples and after ∼50 min of adsorption in the case of quaternized samples. Atomic Force Microscopy (AFM). Changes in surface topography of dried cellulose films with and without adsorbed block copolymers were analyzed with a Nanoscope IIIa multimode scanning probe microscope (Digital Instruments, Santa Barbara, CA). Surface areas of 5 × 5 and 1 × 1 μm2 were scanned in air using tapping mode with silicon cantilevers; at least three separate areas on each sample were imaged. Flattening was the only image processing step applied. X-ray Photoelectron Spectroscopy (XPS). The surface chemical composition of the dried cellulose films with and without adsorbed block copolymers was examined with an AXIS Ultra spectrometer by Kratos Analytical using an X-ray source of monochromatic Al Kα at 100 W irradiation under neutralization. Prior to the measurements, the dried but hygroscopic samples were evacuated overnight. A standard in situ reference (cellulose) was measured with each sample batch in order to monitor experimental conditions. High resolution spectra of carbon (C 1s), oxygen (O 1s), and nitrogen (N 1s) were recorded in

Δf n bare

( )

⎛ ρD2O ⎞ ⎜ ⎟ − 1 ⎝ ρH2O ⎠

ld Δangle 2 m(na − n0)

(2)

where (Δf/n)film is the scaled frequency shift measured when the CNF or TOCNF film is changed from H2O to D2O, (Δf/n)bare is the scaled frequency shift measured when the bare silica sensor is switched from H2O to D2O, and ρH2O and ρD2O are the densities at 20 °C of the solvents (0.9982 and 1.1050 g/cm3 for H2O49 and D2O,50 respectively). The total water content of the film was then calculated according to eq 1. Prior to solvent exchange, CNF films were allowed to swell in Milli-Q water overnight. When a stable baseline at a constant flow of 100 μL/min (T = 20 °C) was reached, the solvent was exchanged to D2O and the frequency shift was monitored. After 10 min, the solvent was changed back to Milli-Q water and the frequency returned back to the initial baseline. This procedure was repeated after in situ TEMPO oxidation of the CNF films. The effect of block copolymer adsorption on the water content of the TOCNF was investigated by studying the H2O/D2O solvent exchange responses before and after the adsorption of the polymer. Surface Plasmon Resonance (SPR). Block copolymer adsorption on TOCNF was monitored in real time with a multiparametric surface plasmon resonance instrument (MPSPR Model Navi 200, Oy BioNavis Ltd., Ylöjärvi, Finland). The SPR effect involves a sharp attenuation of light reflectivity caused by a loss of energy associated with coupling of the directed light with surface plasmon waves traveling along the analyzed solution and solid layer interface. As a result, changes occur in the reflected light energy and angles of maximum reflection intensity.40 The refractive indexes (RI) of media in contact with the surface of the SPR crystal correlate very sensitively with 15278


Article

The Journal of Physical Chemistry B Table 1. Summary of Molecular Structure Characteristics of Poly(2-(dimethylamino) ethyl methacrylate)-blockPoly(oligo(ethylene glycol) methyl methacrylate) (PDMAEMA-b-POEGMA) Copolymersa weight fraction

polymerization degree

sample code

PDMAEMA

POEGMA

DMAEMA

OEGMA

Mncalc (Da)

MnGPC (Da)

PDI

D33-EGMA52 D33-EGMA94 D33-EGMA137 D58-EGMA10 D58-EGMA62 D58-EGMA118 D74-EGMA17 D74-EGMA54 D74-EGMA118

17 10 7 66 24 14 59 31 17

83 90 93 34 76 86 41 69 83

33 33 33 58 58 58 74 74 74

52 94 137 10 62 118 17 54 118

29900 49800 70300 13860 38560 65156 19690 37268 67668

19000 36600 58400 17100 35900 49200 17100 31400 51300

1.21 1.13 1.29 1.20 1.25 1.23 1.17 1.27 1.29

a

Besides the respective polymer sample code, the data include the weight fraction, polymerization degree, calculated molecular weight based on MALDI-TOF and 1H NMR (Mncalc), as well as molecular weight based on GPC (MnGPC) and respective polydispersity indices.

Figure 1. AFM height images 1 × 1 μm2 of regenerated cellulose, CNF, and TOCNF surfaces before (a, d, g) and after adsorption of D33-EGMA137 (b, e, h) and D74-EGMA17 (c, f, i) block copolymers (as indicated on the left). The roughness profiles are included in the bottom of the respective image. The subscript in the polymer descriptor refers to the size of PDMAEMA (D) and POEGMA (EGMA) segments.

different charge density. Figure 1 includes 1 × 1 μm2 AFM images of D33-EGMA137 and D74-EGMA17 adsorbed from aqueous solution on films of regenerated cellulose, CNF, and TOCNF followed by air drying (Figure 1a−c, d−f, and g−i, respectively). Images of the respective bare films are included as reference. The AFM images revealed that the CNF and TOCNF reference, bare films (Figure 1d,g) were already initially much rougher than the regenerated cellulose reference (RMS roughness of 3.4, 3.9 and 0.68 nm, respectively). This is due to the fibrillar network structure of CNF. Differences in AFM topography (as determined by RMS roughness) of regenerated cellulose without and with adsorbed block copolymers are rather negligible (Figure 1a−c): the RMS roughness for regenerated

addition to elemental wide-region data. The analysis area was 400 × 800 μm2. Sample degradation was not observed during measurements. Detailed descriptions on experimental parameters and high-resolution carbon fits are available elsewhere.55

■

RESULTS AND DISCUSSION Block Copolymer Samples. The nine block copolymers synthesized for varying block sizes and comprising PDMAEMA (D) and POEGMA (EGMA) were analyzed by MALDI-TOF, 1 H NMR, and GPC (Table 1). Adsorption of PDMAEMA-b-POEGMA on Cellulose. The nine PDMAEMA-b-POEGMA copolymer samples with varying block sizes were adsorbed onto the cellulose films with 15279


Article


Figure 2. Real time QCM-D shift in frequency, −ΔF3 (a, c), and dissipation, −ΔD3 (b, d), upon D33-EGMA137 and D74-EGMA17 adsorption on regenerated cellulose, CNF, and TOCNF. The vertical arrows indicate the time at which the block copolymer was introduced in the QCM module, and the droplet symbols indicate rinsing with background solvent (PBS buffer, pH 6.8) at given times after adsorption.

for more conclusive evidence. Here, QCM-D adsorption experiments with PDMAEMA-b-POEGMA block copolymers were carried out on the different cellulose surfaces, with a main emphasis on the adsorbed layer mass and changes in viscoelasticity and water coupling. Thus, AFM imaging before and after adsorption of D33-EGMA137 and D74-EGMA17 on cellulose (regenerated cellulose, CNF, and TOCNF films) (Figure 1) was taken as complementary to QCM-D experiments (frequency and dissipation) presented in Figure 2, which provided more insightful observations in relation to adsorption phenomena. The PDMAEMA block contains tertiary amine groups, with a monomer (DMAEMA) pKa value of ca. 8.4,20 which means that ∼50% of the amine groups of the monomers are protonated at this pH. Furthermore, reported pKa values for PDMAEMA of different molecular weight range between 7.4 and 7.8.20 However, in this work, the tertiary amine groups of PDMAEMA blocks were quaternized (permanent cationic charges) by methylation. As discussed previously, PEG molecules do not have remarkable affinity toward cellulose on their own21 and the block copolymers in fact adsorb minimally on regenerated cellulose (Figure 2a and c). The adsorption was somewhat higher on CNF and substantially higher, and very fast, on TOCNF. As the CNF surfaces were oxidized in situ (inside the QCM-D module) via TEMPO reaction, the traditional methods used to determine surface charge were not applicable; therefore, the charge density of TOCNF films was not accessible. However, such a charge density of cellulose across all experiments involving TOCNF was the same given that the same reaction conditions were used. The pKa value of carboxylic acids is in the range 3.0− 5.5,56 and hence, the TOCNF is negatively charged at the experimental pH (6.8): this suggests the substantially high adsorption observed on TOCNF is the result of electrostatic interactions (between positively charged PDMAEMA and TOCNF). This is consistent with the existing literature regarding adsorption of PDMAEMA-containing block copolymers on anionic mica, silica, and cellulose surfaces. However, other effects

cellulose (0.68 nm, reference) after adsorption of D33-EGMA137 and D74-EGMA17 is 0.73 and 0.80 nm, respectively. The small changes in roughness indicate negligible effects upon adsorption of the block copolymers on regenerated cellulose. Likewise, adsorption of D33-EGMA137 and D74-EGMA17 induced no substantial topographical changes on CNF or TOCNF substrates (Figure 1d−i). Altogether, the results from AFM imaging of dried adsorbed layers indicate uniform block polymer adsorption on all tested cellulose surfaces. The height images of CNF and TOCNF with adsorbed block copolymers in the dry state (Figure 1e,f and h,i) show a fibrillar network structure for both substrates and, also, lower height contrast compared to the respective reference images (Figure 1d and g). This observation in height images of CNF (Figure 1e,f) and TOCNF (Figure 1h,i) carrying the block copolymers can be explained by swelling of the PEG-containing blocks during the relatively humid environment used during AFM imaging. However, the topographical differences between the surfaces are more evident if one compares the RMS values of the respective bare CNF surface (3.4 nm) with those of CNF carrying adsorbed D33-EGMA137 and D74-EGMA17 (RMS = 4.3 and 4.2 nm, respectively). Analysis of the RMS roughness values for TOCNF surfaces reveals a negligible change in TOCNF film surfaces upon copolymer adsorption, likely the result of uniform binding onto the substrate. Comparatively, the CNF surfaces were subjected to non-negligible changes in RMS roughness, which can be explained to result from nonuniform block copolymer adsorption and/or the presence of coiled chains at the interface. Therefore, there is indication that a more uniform adsorption or a flatter/tighter conformation of block copolymers on the TOCNF surface occurred due to the influence of electrostatic interactions with the cationic segments of the block copolymers. This is in agreement with results from a previous study where PEG molecules were grafted onto carboxymethyl cellulose and adsorbed onto CNF films.23 We note, however, that the differences in AFM images were rather small and additional inquiries into conformational changes at the interface are needed 15280


Article


Table 2. XPS Atomic Concentrations (%) and Standard Deviations (from Analysis of Wide Spectra) for D33-EGMA137 and D74EGMA17 Adsorbed on Regenerated Cellulose, CNF, and TOCNF Filmsa atomic concentration (%) reg. cellulose (reference) D33-EGMA137 D74-EGMA17 CNF (reference) D33-EGMA137 D74-EGMA17 TOCNF (reference) D33-EGMA137 D74-EGMA17 Whatman a

C 1s

O 1s

N 1s

Si 2p

60.8 ± 0.5 62.5 ± 1.0 62.5 ± 1.2 42.4 ± 2.5 58.9 ± 1.1 60.1 ± 2.1 43.7 ± 0.9 59.0 ± 0.5 57.9 ± 0.5 60.7 ± 0.6

38.2 ± 0.5 36.8 ± 0.9 36.3 ± 1.4 41.3 ± 0.9 30.6 ± 0.8 27.0 ± 0.7 41.1 ± 0.9 30.1 ± 0.6 26.4 ± 0.5 39.3 ± 0.6

0.6 ± 0.5 0.6 ± 0.5 0.7 ± 0.5 1.6 ± 0.5 5.3 ± 0.8 8.3 ± 1.1 0.9 ± 0.5 7.2 ± 0.5 13.1 ± 1.0 N/A

0.1 ± 0.5 N/A 0.3 ± 0.5 14.9 ± 1.6 5.2 ± 0.6 4.3 ± 0.5 14.3 ± 0.5 3.7 ± 0.5 2.5 ± 0.9 N/A

Values for reference films are also included as well as a cellulose standard (Whatman filter paper).

be uniform in AFM (Figure 1d and g), the silicon signal from the substrate was observed for these samples due to contribution from the areas underneath the fibrillar network structure of CNF and TOCNF. XPS analyses are discussed next to determine changes in the composition of the surface upon copolymer adsorption. The nitrogen signal was used as a fingerprint to determine the extent of adsorption for the block copolymers on cellulose. XPS analysis of the regenerated cellulose films before and after block copolymer adsorption supports the observation of little or no adsorption of the block copolymers on regenerated cellulose: the nitrogen surface concentration remains practically unchanged after adsorption of the polymer. The XPS analysis indicates a small contribution of nitrogen originating from the PEI adsorbed on the crystals, which was used as an anchoring layer for CNF and TOCNF. Block copolymer adsorption on CNF and TOCNF is confirmed by increasing nitrogen content. The excess nitrogen can no longer originate from PEI, since the substrate silicon signal is significantly reduced after adsorption. For the CNF films, the amount of nitrogen was clearly increased after adsorption of the block copolymer (from 1.6 (reference) to 5.3 and 8.3% for D33-EGMA137 and D74-EGMA17, respectively). Moreover, the small standard deviations in the nitrogen concentration further support the assumption of uniform block copolymer adsorption on the cellulosic surfaces. Note the somewhat higher N% value for the D74-EGMA17 sample, which contains a large PDMAEMA block. Compared to the TOCNF reference (0.9%), the amount of nitrogen on the TOCNF films was substantially higher upon adsorption of the block copolymers: 7.2% for D33-EGMA137 and 13.1% for D74-EGMA17 (see Supporting Information S3 for XPS spectra for TOCNF with adsorbed D33-EGMA137 and D74EGMA17 block copolymers). The effect of the larger PDMAEMA segment was again seen as a higher N content for the D74EGMA17 adsorbed on TOCNF as compared to D33-EGMA137. The XPS results are in agreement with the QCM-D data, indicating little adsorption of block copolymers on regenerated cellulose, somewhat higher adsorption for CNF films, and a significant adsorption on TOCNF. Table 3 presents the QCM-D frequency and dissipation changes as well as estimated adsorbed masses by Voigt viscoelastic modeling for the nine PDMAEMA-b-POEGMA copolymers adsorbed onto the different cellulose surfaces tested. In most cases, the block copolymer adsorption was almost double on CNF relative to regenerated cellulose films (Table 3). It is also apparent that the adsorbed amounts of block

may not be discounted in these cases, which involved DMAEMA copolymerized with hydrophobic blocks.18,19 The slightly higher adsorption of the block copolymers on CNF compared to regenerated cellulose (Figure 2a and c) can be attributed to the slightly anionic nature of CNF at the adsorption pH (pKa ∼ 4.836), resulting from the residual heteropolysaccharides (mainly xylans) that remain on the fibril surfaces from the precursor fibers (delignification and limited removal of hemicelluloses).56 However, it is also observed that, compared to the case of D74-EGMA17, D33-EGMA137 adsorbed more extensively on CNF, which highlights the role of the POEGMA segment size and the inherent amount of water coupling. Noteworthy, the reduction in QCM dissipation values measured upon block copolymer adsorption on TOCNF (Figure 2b and d) indicates that the interface became more rigid. All of the cellulose surfaces investigated experienced substantial swelling in the aqueous environment, and therefore, a stable QCM-D (frequency and dissipation) baseline was ensured before polymer adsorption. This was carried out by allowing the cellulose system to fully swell (overnight equilibration in water and in the QCM-D module). The TOCNF surfaces were therefore fully hydrated prior to QCM-D measurements. The large reduction in dissipation was most likely due to substantial amounts of water released from the TOCNF via charge neutralization. This large water removal was found to cause peculiar frequency changes: adsorption of some of the block copolymers produced an unexpected positive shift in frequency. Since QCM-D measurements are affected by water coupling, the SPR technique was used to shed further light on the process of block copolymer adsorption on TOCNF. As will be introduced later, SPR measurements for all tested block copolymers indicated their irreversible adsorption on TOCNF surfaces. Therefore, it can be concluded that the positive shift in the QCM frequency that was observed is the result of considerable water expulsion from the TOCNF interface, which is driven by charge neutralization during adsorption. In order to study surface chemical composition, XPS measurements were performed on dried regenerated cellulose, CNF, and TOCNF films, before and after adsorption of D33EGMA137 and D74-EGMA17 block copolymers. The XPS wide atomic concentrations for these samples are presented in Table 2 (see Supporting Information S2 for high resolution carbon data). Both wide and high resolution spectra included the typical cellulose signatures. 55 According to the XPS data, the regenerated cellulose films covered the silicon substrate almost completely. Although the CNF and TOCNF films were found to 15281


Article

The Journal of Physical Chemistry B Table 3. QCM-D Frequency and Dissipation Changes upon Adsorption of PDMAEMA-b-POEGMA Copolymers with Differing Block Sizes on Regenerated Cellulose, CNF, and TOCNF Filmsa block copolymer D33-EGMA52

D33-EGMA94

D33-EGMA137

D58-EGMA10

D58-EGMA62

D58-EGMA118

D74-EGMA17

D74-EGMA54

D74-EGMA118

surface

ΔF3 (Hz)

ΔD3 (1 × 10−6)

Δm (mg/m2)

reg. cellulose CNF TOCNF reg. cellulose CNF TOCNF reg. cellulose CNF TOCNF reg. cellulose CNF TOCNF reg. cellulose CNF TOCNF reg. cellulose CNF TOCNF reg. cellulose CNF TOCNF reg. cellulose CNF TOCNF reg. cellulose CNF TOCNF

−8.4 −15.0 86.3 −5.3 −19.3 −177.5 −3.3 −20.8 −101.6 −5.2 −8.0 −146.2 −4.3 −18.2 77.4 −4.6 −14.8 −7.2 −5.5 −8.8 −140.0 −3.6 −12.9 92.5 −6.4 −23.7 −122.3

1.2 0.8 −28.4 0.8 1.6 −46.5 0.4 1.5 −28.1 0.6 −0.1 12.5 0.7 0.7 −21.4 1.0 0.5 −9.6 1.6 0.3 7.3 −3.6 −0.1 −37.7 1.0 1.6 −7.7

1.6 ± 1.1 2.9 ± 0.4 N/A 1.1 ± 0.1 3.4 ± 0.2 N/A 0.6 ± 0.2 3.6 ± 0.1 N/A 1.0 ± 0.4 1.6 ± 0.4 N/A 1.3 ± 0.1 3.5 ± 0.3 N/A 1.5 ± 0.4 2.5 ± 0.9 N/A 1.3 ± 0.3 1.6 ± 0.4 N/A 1.3 ± 0.1 2.8 ± 0.2 N/A 1.5 ± 0.9 3.8 ± 1.5 N/A

Figure 3. QCM-D third overtone frequency sensogram upon solvent exchange, from H2O to D2O and back to H2O, for bare silica crystal and for a crystal with spin-coated CNF film. Dashed vertical lines indicate the solvent exchange. Double-ended arrows indicate the frequency shifts that were used to calculate the amount of water in the layer by eqs 1 and 2.

H2O/D2O solvent exchange experiments were performed for two block copolymers: D33-EGMA137 and D74-EGMA54 (see Supporting Information S4). These block copolymers were chosen as D33-EGMA137 exhibited significant negative changes in the QCM frequency while D74-EGMA54 produced substantial positive shifts of QCM frequency. The amount of coupled water in the CNF and TOCNF films was determined to be 60 ± 11 and 63 ± 15%, respectively. Rather surprisingly, TEMPO-oxidation did not significantly increase the amount of water coupling in the cellulose layer, which has been observed when the charge density of CNF is increased with adsorption of carboxymethyl cellulose.35 The somewhat high variation is caused by the substantial changes in the CNF mass deposited by spin-coating; nevertheless, the results correlate fairly well with previously published data on water content in CNF film.35 The amount of bound water in the TOCNF layer remained unchanged upon D33-EGMA137 adsorption, while in the case of D74-EGMA54 a very minor reduction in the amount of coupled water (∼4%) was observed. These results can be explained by the decrease in coupled water for block copolymer with the longer cationic segments, thus presenting a higher capacity for charge neutralization. The observed changes are much lower than observed for CNF films and bulk fibers upon cationic polyelectrolyte adsorption (∼10−20%);35,58 water binds extensively to the highly hydrophilic and uncoiled PEG chains of the block copolymers, and they compensate for the water removal caused by charge neutralization by anionic CNF and cationic blocks. The evidence of interfacial water expulsion upon adsorption of PDMAEMA-b-POEGMA on TOCNF was further investigated with SPR. SPR measures the adsorbed layer thickness and mass by determination of the layer refractive index, which is not affected by changes in water coupling. The QCM-D and SPR data upon adsorption on TOCNF of selected PDMAEMA-bPOEGMA copolymers is presented in Figure 4. Depending on the polymer architecture, PDMAEMA− POEGMA block copolymers exhibited either positive or negative shifts in QCM frequency upon adsorption (Figure 4a). Such shifts not only are related to the mass update but are also affected by charge neutralization and water expulsion from the interface (dehydration). In contrast, SPR measurements (Figure 4b) confirmed that all block copolymers in fact adsorbed to a significant extent onto TOCNF. Thus, the adsorption behavior of the block copolymers observed in QCM-D, as anticipated from previous discussions, is indeed a result of water removal

a The adsorbed mass (Δm) was obtained by the Voigt viscoelastic model. The amounts of block copolymer adsorbed on TOCNF are not included due to the effect of water removal, as discussed in the text.

copolymers on regenerated cellulose are approximately the same, ∼1 mg/m2, regardless of the sizes of the PDMAEMA and POEGMA blocks. The adsorption of the block copolymers on CNF films increased with the length of the POEGMA segment. This is explained by the increased POEGMA segment molecular weight, which causes an increase in the amount of bound water, which is detected in QCM-D as a large adsorbed mass. Larsson et al. reported similar results for PDMAEMA−poly(di(ethylene glycol) methyl ether methacrylate) (PDEGMA) block copolymer adsorbed on carboxymethylated CNF.57 Expulsion of Coupled Water upon PDMAEMA-bPOEGMA Adsorption. The changes in the amount of coupled water upon block copolymer adsorption on TOCNF films were further explored with H2O/D2O solvent exchange.35,38 The QCM-D sensograms corresponding to bare silica crystal and silica crystal carrying spin-coated cellulose nanofibrils are presented in Figure 3. The mass of the fibril layer can be calculated (eq 1) by comparing the fundamental resonance of the silica crystal in air before and after CNF spin-coating. The deposition (spin-coating) of CNF causes a decrease in the fundamental resonance of the crystal, corresponding to a CNF areal mass of 10.2 ± 4.4 mg/m2. The substantial variation in the calculated mass (standard deviation of ca. 50%) has also been previously observed for similar spin-coated CNF films and can be attributed to the heterogeneity of the CNF suspension.35 15282


Article


Figure 4. QCM-D (a) and SPR (b) sensograms for adsorption on TOCNF of PDMAEMA-b-POEGMA copolymers with varying segment length. The arrows in parts a and b indicate the point when adsorption with respective block copolymer was initiated; droplet symbols indicate rinsing after adsorption plateau with respective buffer solution (pH 6.8).

interactions between the TOCNF surface and PDMAEMA blocks were not screened substantially. The positively charged PDMAEMA blocks adsorbed tightly and in a flat conformation on the anionic surface, while the POEGMA blocks extended to the aqueous medium, as previously observed for poly(L-lysine)− poly(ethylene glycol) graft-copolymers.60 Therefore, more block copolymers with short PDMAEMA segments fitted the adsorbed layer on the surface compared to block copolymers with longer cationic segments. Also, the large number of POEGMA units in the block copolymer, for a given PDMAEMA block size, were sensed as a large adsorbed mass (Table 4). This is due to the increased mass detected, with highly hydrated polymer chains on the surface of the TOCNF film.57 It should be noted that, in the case of block copolymers with 33 or 58 DMAEMA monomers, the adsorbed amount first increased with the length of the POEGMA segment (case of D33-EGMA94 and D58-EGMA62) to then decrease (case D33-EGMA137 and D58-EGMA118). This is in contrast to the monotonic adsorbed mass increase observed for increased POEGMA segment size for block copolymers carrying 74 DMAEMA monomers. However, the effect highlighted before for block copolymers with 33 or 58 DMAEMA monomers is rather limited and within the reported experimental errors. The heterogeneity of the spin-coated substrate material may be a reason for any unexpected behavior. Effect of PDMAEMA−POEGMA Copolymer Structure on Adsorption on TEMPO-Oxidized CNF. The SPR sensograms for adsorption on TOCNF of PDMAEMA-rndPOEGMA copolymers with given molecular ratios, D46EGMA139 (10−90 mol %), D86-EGMA67 (30−70 mol %), D159EGMA23 (70−30 mol %), and D222EGMA7 (92−8 mol %) (see Supporting Information S5), both under charged (pH ∼ 6.8) and uncharged (pH 5.5 and pH 9.5) conditions, are presented in Figure 5. From this figure, it is apparent that for all random copolymers adsorption was substantially increased by quaternization. The adsorption isotherms for the charged random copolymers are quite different compared to typical Langmuir adsorption kinetics (Figure 5): after very rapid, initial adsorption, the amount of quaternized polymer levels off with a gradual (but linearly increasing) adsorption rate. It can be assumed that the initial fast adsorption is caused by electrostatic interaction between the quaternized DMAEMA units and the anionic TOCNF. After this, the adsorption rate decreases possibly due to multilayer formation, polymer reconformation on the surface, polymer exchange (between copolymers of different sizes due to their polydispersity), or slow counterion release from the quaternized copolymer layer. In general, adsorption of the charged random copolymers increases when the molecular weight of the polymer and the ratio of OEGMA

from the interface, which leads to otherwise unexpected frequency shifts (reduction in frequency). The adsorption of PDMAEMA-b-POEGMA copolymers on TOCNF monitored with SPR (Figure 4b) was very rapid, as the adsorption plateau was reached in a few minutes; rinsing did not remove significant amounts of the copolymers. This is a typical phenomenon for molecules adsorbing by electrostatic interactions.59 Comparing the adsorption of D33-EGMA52 and D74EGMA54, it is evident that adsorption onto TOCNF increases with the size of the PDMAEMA block, likely through more pronounced electrostatic interaction and as a result of the increased entropy as counterions on the surface are displaced by the adsorbing polymer chains. Close comparison of adsorption data for D74-EGMA17 and D74-EGMA118 reveals an increase with the size of the hydrophilic block. This is similar to the interpretation made from the QCM-D results for block copolymer adsorption on CNF: the adsorbed mass is increased possibly due to the adsorbing PDMAEMA segment, which anchors the rest of the block copolymer with large, dangling POEGMA blocks that bind water extensively.57 The layer thicknesses calculated from SPR experiments (eq 3) and the adsorbed amount (eq 4) for all nine polymer samples are given in Table 4. Table 4. Layer Thicknesses and Adsorbed Amounts Calculated from SPR Measurements (eqs 3 and 4) with Standard Deviations for PDMAEMA-b-POEGMA Copolymers Adsorbed on TOCNF block copolymer

d (nm)

Δm (mg/m2)

D33-EGMA52 D33-EGMA94 D33-EGMA137 D58-EGMA10 D58-EGMA62 D58-EGMA118 D74-EGMA17 D74-EGMA54 D74-EGMA118

2.6 ± 0.3 5.9 ± 0.3 5.3 ± 0.6 2.5 ± 0.6 7.0 ± 0.4 5.6 ± 1.2 4.0 ± 0.4 5.4 ± 0.1 7.8 ± 0.2

3.0 ± 0.3 6.8 ± 0.4 6.1 ± 0.6 2.9 ± 0.7 8.1 ± 0.5 6.4 ± 1.4 4.6 ± 0.4 6.2 ± 0.1 8.9 ± 0.2

The results presented in Table 4 confirm the fact that copolymer adsorption onto TOCNF increases as the size of the PDMAEMA segments increase, due to more prominent electrostatic interactions. The adsorbed amounts of D33EGMA137 and D33-EGMA94 were somewhat high compared to the adsorbed amounts of copolymers with 58 or 74 DMAEMA monomers. Since the ionic strength of the phosphate buffer solution (2.5 mM) was relatively low, the electrostatic 15283


Article


Figure 5. Adsorption on TOCNF of quaternized (pH ∼ 6.8) and uncharged (pH 5.5 and 9.5) PDMAEMA-rnd-POEGMA copolymers monitored by SPR, as indicated. The arrows symbolize the time when the random copolymer injection was initiated, and the droplet symbols correspond to rinsing with the respective buffer solution.

units is increased: 20.1 ± 1.9 and 12.1 ± 0.9 mg/m2 for D46EGMA139 (73 kDa, 10−90 mol %) and D159EGMA23 (36 kDa, 70−30 mol %), respectively (Supporting Information S6 and S7). This is again due to the higher molecular mass of OEGMA compared to DMAEMA monomer: as the molar mass of the copolymer increases, a higher adsorbed mass is detected.57 Interestingly, the adsorption kinetics as well as the adsorbed amount of quaternized PDMAEMA-rnd-POEGMA on TOCNF are rather different from that of the respective block copolymers. For all random copolymers after fast initial adsorption, a gradual increasing adsorption is observed (Figure 5), whereas for the block copolymers the adsorption plateau is reached very fast in a few minutes (Figure 4b). As discussed before, the cationic segments of the block copolymers adsorb tightly and flat on the TOCNF surface while the POEGMA chains extend into the solution. However, in the case of the random copolymers, the random DMAEMA segments/monomers on the polymer adsorb on the TOCNF surface while the OEGMA segments extend to the solution, forming loops and tails. Comparing the adsorbed amounts for random copolymer D46EGMA139 (20.1 ± 1.9 mg/m2) and block copolymer D33EGMA137 (6.1 ± 0.6 mg/m2), which have similar molecular ratios and molecular mass (10−90 and 7−93 mol %, and 70 kDa, respectively), it is clear that adsorption on TOCNF is significantly higher for the random copolymer. Overall, the molecular weights of the random copolymers (Supporting Information S5) were somewhat higher than those of the respective block copolymers (Table 1), which can account to some extent for the higher adsorption observed in SPR. However, it is evident that the high adsorption and peculiar adsorption kinetics of the PDMAEMA-rnd-POEGMA copolymers on TOCNF still require further investigation. Nonquaternized copolymers were adsorbed on TOCNF films from acetate (pH 5.5) and carbonate (pH 9.5) buffer solutions (Figure 5), thus allowing us to elucidate the effect of the DMAEMA on adsorption (since it is positively charged when the

tertiary amine groups of DMAEMA monomers are protonated in pH 5.5 or deprotonated in pH 9.5 (pKa 7.4−7.8)).20 The positive charge of DMAEMA monomers is the main driving force for adsorption on anionic TOCNF, since, as observed, adsorption was significantly higher at pH 5.5. At pH 9.5, adsorption was rather small and of similar magnitude for all random copolymers. The increased amount of OEGMA, as before, is observed to increase adsorption at pH 5.5, similarly as for the charged polymer. If one compares these results for the adsorption of quaternized PDMAEMA−POEGMA and those of random (Figure 5) and block (Figure 4b) polymers, it is evident that quaternization increases adsorption onto TOCNF.

■

CONCLUSIONS Quaternized PDMAEMA-b-POEGMA copolymer adsorption on regenerated cellulose, CNF, and TOCNF films was studied. By using complementary QCM-D and SPR data, it was determined that block copolymer adsorption on TOCNF and CNF was mainly driven by electrostatic interactions while the increased size of the POEGMA segment had a favorable effect on the sensed adsorbed mass. Very little adsorption occurred on regenerated cellulose films, due to the absence of electrostatic interactions. XPS analysis of the surfaces after block copolymer adsorption further confirms that PDMAEMA-b-POEGMA adsorbs to the largest extent on TOCNF and very limited on regenerated cellulose. AFM images revealed that the block copolymers were uniformly attached to the surface, making the fibrillar network features less distinctive. The block copolymers were found to cause positive and negative QCM-D frequency shifts upon adsorption onto TOCNF, while SPR measurements confirmed a significant adsorption for all block copolymers. Therefore, the removal of interfacial water and their effect upon polymer adsorption is clearly proven. The amount and kinetics of adsorption of similar but random quaternized PDMAEMA−POEGMA copolymers on TOCNF was found to differ substantially from those observed 15284


Article


(6) Metreveli, G.; Wågberg, L.; Emmoth, E.; Belák, S.; Strømme, M.; Mihranyan, A. A Size Exclusion Nanocellulose Filter Paper for Virus Removal. Adv. Healthcare Mater. 2014, 3, 1546−1550. (7) Arola, S.; Tammelin, T.; Setälä, H.; Tullila, A.; Linder, M. B. Immobilization - Stabilization of Proteins on Nanofibrillated Cellulose Derivatives and their Bioactive Film Formation. Biomacromolecules 2012, 13, 594−603. (8) Andresen, M.; Stenstad, P.; Møretrø, T.; Langsrud, S.; Syverud, K.; Johansson, L.; Stenius, P. Nonleaching Antimicrobial Films Prepared from Surface-Modified Microfibrillated Cellulose. Biomacromolecules 2007, 8, 2149−2155. (9) Syverud, K.; Stenius, P. Strength and Barrier Properties of MFC Films. Cellulose 2009, 16, 75−85. (10) Abe, K.; Iwamoto, S.; Yano, H. Obtaining Cellulose Nanofibers with a Uniform Width of 15 Nm from Wood. Biomacromolecules 2007, 8, 3276−3278. (11) Päak̈ kö, M.; Ankerfors, M.; Kosonen, H.; Nykänen, A.; Ahola, S.; Ö sterberg, M.; Ruokolainen, J.; Laine, J.; Larsson, P. T.; Ikkala, O.; Lindström, T. Enzymatic Hydrolysis Combined with Mechanical Shearing and High-Pressure Homogenization for Nanoscale Cellulose Fibrils and Strong Gels. Biomacromolecules 2007, 8, 1934−1941. (12) Saito, T.; Kimura, S.; Nishiyama, Y.; Isogai, A. Cellulose Nanofibers Prepared by TEMPO-Mediated Oxidation of Native Cellulose. Biomacromolecules 2007, 8, 2485−2491. (13) Liu, X.; He, F.; Salas, C.; Pasquinelli, M. A.; Genzer, J.; Rojas, O. J. Experimental and Computational Study of the Effect of Alcohols on the Solution and Adsorption Properties of a Nonionic Symmetric Triblock Copolymer. J. Phys. Chem. B 2012, 116, 1289−1298. (14) Liu, H.; Li, Y.; Krause, W. E.; Pasquinelli, M. A.; Rojas, O. J. Mesoscopic Simulations of the Phase Behavior of Aqueous EO19PO29EO19 Solutions Confined and Sheared by Hydrophobic and Hydrophilic Surfaces. ACS Appl. Mater. Interfaces 2012, 4, 87−95. (15) Li, Y.; Liu, H.; Song, J.; Rojas, O. J.; Hinestroza, J. P. Adsorption and Association of a Symmetric PEO-PPO-PEO Triblock Copolymer on Polypropylene, Polyethylene, and Cellulose Surfaces. ACS Appl. Mater. Interfaces 2011, 3, 2349−2357. (16) Liu, X.; Wu, D.; Turgman-Cohen, S.; Genzer, J.; Theyson, T. W.; Rojas, O. J. Adsorption of a Nonionic Symmetric Triblock Copolymer on Surfaces with Different Hydrophobicity. Langmuir 2010, 26, 9565− 9574. (17) Liu, X.; Vesterinen, A.; Genzer, J.; Seppälä, J. V.; Rojas, O. J. Adsorption of PEO−PPO−PEO Triblock Copolymers with EndCapped Cationic Chains of Poly (2-Dimethylaminoethyl Methacrylate). Langmuir 2011, 27, 9769−9780. (18) Nurmi, L.; Kontturi, K.; Houbenov, N.; Laine, J.; Ruokolainen, J.; Seppälä, J. Modification of Surface Wettability through Adsorption of Partly Fluorinated Statistical and Block Polyelectrolytes from Aqueous Medium. Langmuir 2010, 26, 15325−15332. (19) Wang, M.; Olszewska, A.; Walther, A.; Malho, J.; Schacher, F. H.; Ruokolainen, J.; Ankerfors, M.; Laine, J.; Berglund, L. A.; Osterberg, M. Colloidal Ionic Assembly between Anionic Native Cellulose Nanofibrils and Cationic Block Copolymer Micelles into Biomimetic Nanocomposites. Biomacromolecules 2011, 12, 2074−2081. (20) Van de Wetering, P.; Zuidam, N.; Van Steenbergen, M.; Van der Houwen, O.; Underberg, W.; Hennink, W. A Mechanistic Study of the Hydrolytic Stability of Poly (2-(Dimethylamino) Ethyl Methacrylate). Macromolecules 1998, 31, 8063−8068. (21) Filpponen, I.; Kontturi, E.; Nummelin, S.; Rosilo, H.; Kolehmainen, E.; Ikkala, O.; Laine, J. Generic Method for Modular Surface Modification of Cellulosic Materials in Aqueous Medium by Sequential “click” Reaction and Adsorption. Biomacromolecules 2012, 13, 736−742. (22) Wang, S.; Guo, X.; Wang, L.; Wang, W.; Yu, Y. Effect of PEG Spacer on Cellulose Adsorbent for the Removal of Low Density Lipoprotein-Cholesterol. Artif. Cells, Blood Substitutes, Biotechnol. 2006, 34, 101−112. (23) Olszewska, A.; Junka, K.; Nordgren, N.; Laine, J.; Rutland, M. W.; Ö sterberg, M. Non-Ionic Assembly of Nanofibrillated Cellulose and Polyethylene Glycol Grafted Carboxymethyl Cellulose and the Effect of

for respective block copolymers. Compared to quaternized PDMAEMA-b-POEGMA, the quaternized, random copolymer samples adsorbed more significantly onto TOCNF.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcb.5b07628. In situ TEMPO oxidation of CNF films in QCM-D and SPR, XPS high resolution carbon spectra for regenerated cellulose, CNF, and TOCNF films before and after adsorption of D33-EGMA137 and D74-EGMA17, H2O/D2O solvent exchange QCM-D sensograms for bare silica sensors, cellulose, CNF, and TOCNF films with adsorbed D33-EGMA137 or D74-EGMA54, characterization information for analyzed PDMAEMA-rnd-POEGMA copolymers with given molecular ratios, calculated PDMAEMA-rndPOEGMA copolymer layer thickness, and adsorbed mass on TOCNF (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +358 40 3543143. E-mail: hannes.orelma@vtt.fi. Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors are grateful for funding support by the Academy of Finland through its Centres of Excellence Programme (2014− 2019), under Project 264677 “Molecular Engineering of Biosynthetic Hybrid Materials Research” (HYBER) and also under the project “Nanocellulose Assemblies for Microsensing and Fluidics” as well as by the WoodWisdom-Net project “SelfAssembled Biomimetic Wood-Based Nanocomposites (CellAssembly)”. A.W. acknowledges funding from the BMBF in the framework of the Aquamat research group. Prof. Olli Ikkala is acknowledged for valuable comments and Dr. Joseph Campbell for performing XPS measurements. Ritva Kivelä and Anu Anttila are greatly thanked for their technical assistance.

■

REFERENCES

(1) Eichhorn, S.; et al. Review: Current International Research into Cellulose Nanofibres and Nanocomposites. J. Mater. Sci. 2010, 45, 1− 33. (2) Siqueira, G.; Bras, J.; Dufresne, A. Cellulose Whiskers Versus Microfibrils: Influence of the Nature of the Nanoparticle and its Surface Functionalization on the Thermal and Mechanical Properties of Nanocomposites. Biomacromolecules 2009, 10, 425−432. (3) Iwamoto, S.; Isogai, A.; Iwata, T. Structure and Mechanical Properties of Wet-Spun Fibers made from Natural Cellulose Nanofibers. Biomacromolecules 2011, 12, 831−836. (4) Turbak, A. F.; Snyder, F. W.; Sandberg, K. R. Microfibrillated Cellulose, a New Cellulose Product: Properties, Uses, and Commercial Potential. J. Appl. Polym. Sci.: Appl. Polym. Symp. 1983, 37, 815−827. (5) Jiang, F.; Hsieh, Y. Super Water Absorbing and Shape Memory Nanocellulose Aerogels from TEMPO-Oxidized Cellulose Nanofibrils Via Cyclic Freezing−thawing. J. Mater. Chem. A 2014, 2, 350−359. 15285


Article

The Journal of Physical Chemistry B Aqueous Lubrication in Nanocomposite Formation. Soft Matter 2013, 9, 7448−7457. (24) Braithwaite, G.; Howe, A.; Luckham, P. Interactions between Poly (Ethylene Oxide) Layers Adsorbed to Glass Surfaces Probed by using a Modified Atomic Force Microscope. Langmuir 1996, 12, 4224−4237. (25) Maddikeri, R.; Tosatti, S.; Schuler, M.; Chessari, S.; Textor, M.; Richards, R.; Harris, L. Reduced Medical Infection Related Bacterial Strains Adhesion on Bioactive RGD Modified Titanium Surfaces: A First Step Toward Cell Selective Surfaces. J. Biomed. Mater. Res., Part A 2008, 84, 425−435. (26) Yan, X.; Perry, S. S.; Spencer, N. D.; Pasche, S.; De Paul, S. M.; Textor, M.; Lim, M. S. Reduction of Friction at Oxide Interfaces upon Polymer Adsorption from Aqueous Solutions. Langmuir 2004, 20, 423− 428. (27) Drobek, T.; Spencer, N. D. Nanotribology of Surface-Grafted PEG Layers in an Aqueous Environment. Langmuir 2008, 24, 1484− 1488. (28) Lee, S.; Müller, M.; Heeb, R.; Zürcher, S.; Tosatti, S.; Heinrich, M.; Amstad, F.; Pechmann, S.; Spencer, N. Self-Healing Behavior of a Polyelectrolyte-Based Lubricant Additive for Aqueous Lubrication of Oxide Materials. Tribol. Lett. 2006, 24, 217−223. (29) Chawla, K.; Lee, S.; Lee, B. P.; Dalsin, J. L.; Messersmith, P. B.; Spencer, N. D. A Novel Low friction Surface for Biomedical Applications: Modification of Poly (Dimethylsiloxane) (PDMS) with Polyethylene Glycol (PEG)-DOPA-lysine. J. Biomed. Mater. Res., Part A 2009, 90, 742−749. (30) VandeVondele, S.; Vörös, J.; Hubbell, J. A. RGD-Grafted Poly-LLysine-Graft-(Polyethylene Glycol) Copolymers Block Non-Specific Protein Adsorption while Promoting Cell Adhesion. Biotechnol. Bioeng. 2003, 82, 784−790. (31) Isogai, A.; Saito, T.; Fukuzumi, H. TEMPO-Oxidized Cellulose Nanofibers. Nanoscale 2011, 3, 71−85. (32) Matthews, J. F.; Skopec, C. E.; Mason, P. E.; Zuccato, P.; Torget, R. W.; Sugiyama, J.; Himmel, M. E.; Brady, J. W. Computer Simulation Studies of Microcrystalline Cellulose Iβ. Carbohydr. Res. 2006, 341, 138−152. (33) Driemeier, C.; Bragatto, J. Crystallite Width Determines Monolayer Hydration Across a Wide Spectrum of Celluloses Isolated from Plants. J. Phys. Chem. B 2013, 117, 415−421. (34) Mohan, T.; Spirk, S.; Kargl, R.; Doliška, A.; Vesel, A.; Salzmann, I.; Resel, R.; Ribitsch, V.; Stana-Kleinschek, K. Exploring the Rearrangement of Amorphous Cellulose Model Thin Films upon Heat Treatment. Soft Matter 2012, 8, 9807−9815. (35) Kontturi, K. S.; Kontturi, E.; Laine, J. Specific Water Uptake of Thin Films from Nanofibrillar Cellulose. J. Mater. Chem. A 2013, 1, 13655−13663. (36) Wågberg, L.; Decher, G.; Norgren, M.; Lindström, T.; Ankerfors, M.; Axnäs, K. The Build-Up of Polyelectrolyte Multilayers of Microfibrillated Cellulose and Cationic Polyelectrolytes. Langmuir 2008, 24, 784−795. (37) Craig, V. S.; Plunkett, M. Determination of Coupled Solvent Mass in Quartz Crystal Microbalance Measurements using Deuterated Solvents. J. Colloid Interface Sci. 2003, 262, 126−129. (38) Kittle, J. D.; Du, X.; Jiang, F.; Qian, C.; Heinze, T.; Roman, M.; Esker, A. R. Equilibrium Water Contents of Cellulose Films Determined Via Solvent Exchange and Quartz Crystal Microbalance with Dissipation Monitoring. Biomacromolecules 2011, 12, 2881−2887. (39) Rodahl, M.; Kasemo, B. A Simple Setup to Simultaneously Measure the Resonant Frequency and the Absolute Dissipation Factor of a Quartz Crystal Microbalance. Rev. Sci. Instrum. 1996, 67, 3238− 3241. (40) Schasfoort, R. B. M., Tudos, A. J., Eds. Handbook of surface plasmon resonance; Royal Society of Chemistry: Cambridge, U.K., 2008. (41) Cooper, G. K.; Sandberg, K. R.; Hinck, J. F. Trimethylsilyl Cellulose as Precursor to Regenerated Cellulose Fiber. J. Appl. Polym. Sci. 1981, 26, 3827−3836. (42) Kontturi, E.; Thüne, P. C.; Niemantsverdriet, J. W. Cellulose Model Surfaces - Simplified Preparation by Spin Coating and Characterization by X-Ray Photoelectron Spectroscopy, Infrared

Spectroscopy, and Atomic Force Microscopy. Langmuir 2003, 19, 5735−5741. (43) Ahola, S.; Salmi, J.; Johansson, L.; Laine, J.; Ö sterberg, M. Model Films from Native Cellulose Nanofibrils. Preparation, Swelling, and Surface Interactions. Biomacromolecules 2008, 9, 1273−1282. (44) Rodahl, M.; Hook, F.; Krozer, A.; Brzezinski, P.; Kasemo, B. Quartz Crystal Microbalance Setup for Frequency and Q-Factor Measurements in Gaseous and Liquid Environments. Rev. Sci. Instrum. 1995, 66, 3924−3930. (45) Sauerbrey, G. The use of Quartz Oscillators for Weighing Thin Layers and for Microweighing. Eur. Phys. J. A 1959, 155, 206−222. (46) Rodahl, M.; Kasemo, B. On the Measurement of Thin Liquid Overlayers with the Quartz-Crystal Microbalance. Sens. Actuators, A 1996, 54, 448−456. (47) Voinova, M. V.; Rodahl, M.; Jonson, M.; Kasemo, B. Viscoelastic Acoustic Response of Layered Polymer Films at Fluid-Solid Interfaces: Continuum Mechanics Approach. Phys. Scr. 1999, 59, 391. (48) Kittle, J. D.; Wondraczek, H.; Wang, C.; Jiang, F.; Roman, M.; Heinze, T.; Esker, A. R. Enhanced Dewatering of Polyelectrolyte Nanocomposites by Hydrophobic Polyelectrolytes. Langmuir 2012, 28, 11086−11094. (49) Zwolinski, B. J.; Eicher, L. D. High-Precision Viscosity of Supercooled Water and Analysis of the Extended Range Temperature Coefficient. J. Phys. Chem. 1971, 75, 2016−2024. (50) Lide, D. R., Ed. Handbook of Chemistry and Physics; CRC Press: Boca Raton, FL, 1992; p 1. (51) Jung, L. S.; Campbell, C. T.; Chinowsky, T. M.; Mar, M. N.; Yee, S. S. Quantitative Interpretation of the Response of Surface Plasmon Resonance Sensors to Adsorbed Films. Langmuir 1998, 14, 5636−5648. (52) Campbell, C. T.; Kim, G. SPR Microscopy and its Applications to High-Throughput Analyses of Biomolecular Binding Events and their Kinetics. Biomaterials 2007, 28, 2380−2392. (53) Lee, B. S.; Chi, Y. S.; Lee, K.; Kim, Y.; Choi, I. S. Functionalization of Poly(Oligo(Ethylene Glycol) Methacrylate) Films on Gold and Si/ SiO2 for Immobilization of Proteins and Cells: SPR and QCM Studies. Biomacromolecules 2007, 8, 3922−3929. (54) Feng, W.; Zhu, S.; Ishihara, K.; Brash, J. Protein Resistant Surfaces: Comparison of Acrylate Graft Polymers Bearing OligoEthylene Oxide and Phosphorylcholine Side Chains. Biointerphases 2006, 1, 50−60. (55) Johansson, L.; Campbell, J. Reproducible XPS on Biopolymers: Cellulose Studies. Surf. Interface Anal. 2004, 36, 1018−1022. (56) Sjöström, E. The Origin of Charge on Cellulosic Fibers. Nord. Pulp Pap. Res. J. 1989, 4, 90−93. (57) Larsson, E.; Sanchez, C. C.; Porsch, C.; Karabulut, E.; Wågberg, L.; Carlmark, A. Thermo-Responsive Nanofibrillated Cellulose by Polyelectrolyte Adsorption. Eur. Polym. J. 2013, 49, 2689−2696. (58) Swerin, A.; Ö dberg, L.; Lindström, T. Deswelling of Hardwood Kraft Pulp Fibres by Cationic Polymers. the Effect on Wet Pressing and Sheet Properties. Nord. Pulp Pap. Res. J. 1990, 5, 188−196. (59) Hoogeveen, N. G.; Stuart, M. A. C.; Fleer, G. J. Polyelectrolyte Adsorption on Oxides: II. Reversibility and Exchange. J. Colloid Interface Sci. 1996, 182, 146−157. (60) Kenausis, G. L.; Vörös, J.; Elbert, D. L.; Huang, N.; Hofer, R.; Ruiz-Taylor, L.; Textor, M.; Hubbell, J. A.; Spencer, N. D. Poly (LLysine)-G-Poly (Ethylene Glycol) Layers on Metal Oxide Surfaces: Attachment Mechanism and Effects of Polymer Architecture on Resistance to Protein Adsorption. J. Phys. Chem. B 2000, 104, 3298− 3309.

15286


Effect of Molecular Architecture of PDMAEMA ... - ACS Publications

Recommend Documents